rosin up, and put on your speculation spectacles

Listening to Unfolding this morning, I was struck by how “strings-like” many of its voices are in general timbre, regardless of what was producing those sounds. I kind of wish I could take credit for brilliant imitative sound design, but honestly: I didn’t actively set out to imitate string ensembles. There’s just something that really works about that kind of sound though.

The basic formula for a synth string ensemble is fairly simple: sawtooth waves, preferably from a stack of detuned oscillators or a chorus effect later, with a lowpass filter to tame it a little, a sustained volume envelope, and typically some reverb. Playing chords helps with the illusion, as does vibrato. To get a little fancier, a fixed filter bank or physical modeling can be employed, or a modulated wavetable that adds a bit more motion. None of this will necessarily fool the ear into thinking it’s actual strings being vibrated by actual horsehair stretched across actual wood, but it can definitely put the sound of a string section in mind.

While I also approximated strings with the Hertz Donut, Lyra-8 and others — thanks mainly to generous chorus and reverb — most of the points on the album that sound most convincingly like a string section or a double bass were Ensemble Oscillator. It’s not really any wonder; the right shapes for it are easy to dial in, the self-FM can add just the right kind of growl and bite so you can practically hear the rosin gripping the strings, and it eats detuned chords for breakfast.

ENOSC of course can do pipe organs with embarrassing ease — so much that when you first try it, it feels difficult to avoid them. It can also go to all kinds of astonishing, unexpected places. You can tune it so its lowest oscillators drop down to LFO rate but set the spread so its highest sing soprano and are modulated by those LFOs. You can coax patterned static bursts and growly flutters and metallic resonant hums and all kinds of oddities.

I am leaning harder toward “I have plenty of oscillators” because the likes of ENOSC, Shapeshifter, and Akemie’s Castle are each just so much in themselves already. And there’s Kermit, Rings, Mimeophon, Angle Grinder, Zorlon, Phonogene, Clouds, Maths etc. to act as sources. And the Medusa, the Microfreak, and the incoming YM2612 synth. And software integrated with the modular.

Mutable Instruments announced a new revision of the Veils quad VCA last week, and also that there will be two more releases. It’s been more than strongly hinted that one of these is the successor to Clouds. The other — which sounds like it’s going to be another successor-with-a-new-name sort of thing — is unknown. So I’ll speculate baselessly!

Edges/Yarns: Edges was a quad chiptune-inspired oscillator, discontinued some time back. Yarns was a MIDI to CV converter that also could act as simple oscillators, discontinued recently due to relatively low sales (and a general decrease in demand for MIDI-CV converters). I could see some kind of quad oscillator with a minijack MIDI input as well as CV perhaps, but I wouldn’t rate it as highly likely.

Peaks: Arguably, this LFO/envelope generator/drum voice was supplanted by Stages as a much more powerful modulation source and Plaits as a more powerful percussion module. With plenty of more basic EG and LFO modules out there I can’t see a compelling reason for a more direct Peaks successor. So I’d rate this as even less likely.

Streams: It could be made non-modal, with a VCA/VCF balance and some integrated compression ability. Maybe not the strongest contender for a remake, but not impossible.

Branches/Grids: in a sense, Marbles does some of this. I honestly don’t have any experience with Grids, but it seems fairly robust already and there’s a lot of other competition out there for pattern sequencing stuff. It’s a possibility, though.

Rings/Elements: my wishful thinking here is that there’s room for another resonator in the Mutable lineup. Perhaps something that emphasizes the pure resonator aspect more, while offering new resonator models (perhaps a continuous morph between models) — or something that leans more to the bowed/breath side as opposed to Rings’ pluck/mallet. Not a “replacement” or “successor” to Rings so much as a complementary module. I can’t even guess if this is likely or just my own daydream though.

Warps: maybe? There are a lot of ways a revision could go and I don’t really have a clear vision of what that would be. I can tell you though, I would go for it.

Frames: I feel like it was a brilliant concept but it suffered somewhat in the implementation. If I were designing my own keyframe-based CV mapping module, it’d have an LCD display on it — but there are other ways in which the interface could be redesigned for clarity. It could also perhaps add individual CV inputs, making it (in some minor sense) a sort of fusion with Blinds, yet remaining at or below 18HP.

Regardless, my plan is now:

  • Continue to wait on the charity auction to see what else comes up.
  • Plan to pick up a Doepfer A-188-1y (256 stage BBD).
  • Reserve the remaining space for Mutable Instruments gear, at least until I see what that’s going to be. And if it’s not a must-have, don’t fill in that space with another oscillator, but consider other options.